**** BEGIN LOGGING AT Fri Dec 03 03:00:00 2010
Dec 03 03:09:38 <kergoth> ntu: what OS/arch?
Dec 03 03:13:28 <ntu> x86_64 on arch linux
Dec 03 04:15:27 <grg> khem, ping?
Dec 03 04:16:46 <grg> a diff between my tree and the release branch shows patches to the following as outstanding: task-proper-tools.bb, gzip_1.4.bb, debianutils.inc, bind, autoconf, automake
Dec 03 04:20:41 <grg> + openjade fix
Dec 03 04:24:34 <grg> i never realised how awesome patchwork was until it stopped working...
Dec 03 06:27:37 <ka6sox> khem, made nuker work.
Dec 03 07:03:07 <eFfeM_work> grg, blindvt have your keys been added already ?
Dec 03 07:04:26 <ka6sox> eFfeM_work, missed him
Dec 03 07:05:10 <eFfeM_work> ah ok, just noticed that he left half an hour ago, and I got just in 10 mins ago
Dec 03 07:23:11 <stefan_schmidt> Mentor grabbed CodeSourcery
Dec 03 07:23:15 <stefan_schmidt> interesting
Dec 03 07:23:33 <stefan_schmidt> http://www.linuxfordevices.com/c/a/News/Mentor-acquires-key-assets-of-CodeSourcery/?kc=rss
Dec 03 07:30:07 <eFfeM_work> stefan_schmidt: seems mentor is quite eager to acquire stuff, iirc they also took over embedded alley a year or so ago
Dec 03 07:30:55 <stefan_schmidt> eFfeM_work: yeah, thats how they came to OE if I got it right
Dec 03 07:31:28 <stefan_schmidt> Time to get Codesourcery customers over to OE  muhahahaha
Dec 03 07:32:02 <eFfeM_work> well if they add some support, that would be nice
Dec 03 07:32:31 <eFfeM_work> nowadays I see more poky/yocto patches than oe patches so we seem to become overtaken
Dec 03 07:39:42 <tasslehoff> I've been sleeping in class, but suddenly noticed the release-2010.12 branch. We're targeting a sw-release in March, and I wonder if I should start following that instead of dev. What's the "idea" behind this branch? 
Dec 03 07:40:31 <stefan_schmidt> tasslehoff: I tested metadata snapshot of the master branch
Dec 03 07:40:50 <stefan_schmidt> tasslehoff: it will get a tag with all successful build combinations
Dec 03 07:41:11 <stefan_schmidt> tasslehoff: So far no group stepped up to make it a long term supported branch
Dec 03 07:41:24 <stefan_schmidt> That means it will only be a tag in git right now.
Dec 03 07:41:43 <stefan_schmidt> But at least one you know was tested more then others and where you base your stuff on
Dec 03 07:42:14 <stefan_schmidt> If a group of interest people gather around it and can decide on a plan to support it it may even get more fixes
Dec 03 07:42:48 <stefan_schmidt> We are trying to do such metadata releases every 3 or 6 months from now to give people a base to work on
Dec 03 07:43:10 <stefan_schmidt> Not all details are decided yet we are still learning how to do it. :)
Dec 03 07:43:16 <stefan_schmidt> tasslehoff: does this help you?
Dec 03 07:43:18 <tasslehoff> stefan_schmidt: cool. thanks for clearing it up for me. :)
Dec 03 07:43:25 <stefan_schmidt> tasslehoff: np
Dec 03 07:43:51 <stefan_schmidt> tasslehoff: So, _maybe_, there will be another release beginning of march
Dec 03 07:44:03 <stefan_schmidt> tasslehoff: if you want to wait or not is your choice :D
Dec 03 07:44:14 <tasslehoff> stefan_schmidt: indeed. I think I might create a local branch starting at the release tag, and merge in the diff between my dev and upstream.
Dec 03 07:45:07 <stefan_schmidt> tasslehoff: thats one option. Depending on what you want to do it may be also an option to collaborate with other on a real release branch on top of the tag
Dec 03 07:48:40 <tasslehoff> stefan_schmidt: I submit patches and report errors when I find oddities, if that's what you mean.
Dec 03 07:49:14 <tasslehoff> stefan_schmidt: so for the release it's the tag I want, not the branch?
Dec 03 07:49:57 <stefan_schmidt> tasslehoff: (patches) thats good. Was not meant as a pun for the local branch. :) OE metadata allows this.
Dec 03 07:50:10 <stefan_schmidt> tasslehoff: the branch will go once we tag the release
Dec 03 07:50:23 <stefan_schmidt> tasslehoff: its only there to hold a subset of patches also in master
Dec 03 07:51:02 <stefan_schmidt> tasslehoff: it will get merged back into master (zero changes, just to let git know) and then a tag is attached for it
Dec 03 07:51:26 <stefan_schmidt> tasslehoff: the tag then hold the name and a matrix of test combinations. Like we are doing for testing
Dec 03 07:52:26 <tasslehoff> stefan_schmidt: sounds good, and also sounds like I will stay on master until the tag is final :)
Dec 03 07:52:48 <stefan_schmidt> tasslehoff: ok
Dec 03 08:23:10 <mckoan> good morning
Dec 03 08:35:50 <hrw> morning
Dec 03 08:36:11 <stefan_schmidt> hi hrw 
Dec 03 08:45:31 <eFfeM_work> hi hrw, ka6sox asked me to tell you that nuke is working
Dec 03 08:49:50 <hrw> eFfeM_work: noticed, used
Dec 03 08:49:59 <eFfeM_work> great
Dec 03 08:52:51 <ericben> good morning
Dec 03 08:53:25 <eFfeM_work> need some git advice, I did a fresh git clone, then did git checkout -b  release-2010.12 origin/release-2010.12; then did git pull but several files are missing, e.g. recipes/perl only has the two perl recipes, but no lib things or perl.inc
Dec 03 08:53:29 <eFfeM_work> what did I miss
Dec 03 08:53:48 <eFfeM_work> ericben: hi & ka6sox also asked me to tell you that nuke is working
Dec 03 08:54:01 <ericben> eFfeM_work: cool but what is nuke ? :-D
Dec 03 08:54:24 <eFfeM_work> a plugin to nuke spammers in the wiki or somehting like that
Dec 03 08:54:31 <ericben> eFfeM_work: ah ok that's a good thing
Dec 03 08:54:51 <ericben> eFfeM_work: I think git pull only pull the origin but doesn't merge it into your branch
Dec 03 08:55:06 <ericben> eFfeM_work: so git rebase origin/release-2010.12 will update your branch
Dec 03 08:55:42 <ericben> or you need to create your branch with --track to track the origin/release-2010.12 branch
Dec 03 08:55:59 <ericben> or do git  branch --set-upstream release-2010.12 origin/release-2010.12
Dec 03 08:56:04 <eFfeM_work> ericben: rebase only gives me 
Dec 03 08:56:05 <eFfeM_work> hudson@ax010-bob64:~/jobs/FM_TEST/workspace/openembedded$ ls recipes/perl
Dec 03 08:56:05 <eFfeM_work> perl_5.10.1.bb	perl_5.8.8.bb
Dec 03 08:56:20 <eFfeM_work> basically I want to start completely fresh and set up a release testing build
Dec 03 08:56:34 <eFfeM_work> but can't seem to get the incantations right
Dec 03 08:56:37 <ericben> eFfeM_work: what says git status
Dec 03 08:56:54 <ericben> it seems a checkout failed 
Dec 03 08:57:05 <eFfeM_work> yeah, lots of files are deleted
Dec 03 08:57:08 <ericben> e
Dec 03 08:57:10 <ericben> ok
Dec 03 08:57:22 <eFfeM_work> how can I force a re-checkout ?
Dec 03 08:57:25 <ericben> to git checkout origin/master; git branch -D release-2010.12 and start again ;-)
Dec 03 08:57:38 <ericben> or git reset --hard HEAD 
Dec 03 08:59:20 <eFfeM_work> reset --hard HEAD worked, but now I lost the branch, 
Dec 03 08:59:35 <ericben> strange
Dec 03 08:59:53 <eFfeM_work> hudson@ax010-bob64:~/jobs/FM_TEST/workspace/openembedded$ git branch
Dec 03 08:59:53 <eFfeM_work> * (no branch)
Dec 03 08:59:53 <eFfeM_work>   master
Dec 03 09:00:05 <eFfeM_work> git and I do not always go together well
Dec 03 09:00:07 <ericben> git so the branch was not completly checked
Dec 03 09:00:11 <ericben> git checkout master
Dec 03 09:00:33 <ericben> then git checkout -b  release-2010.12 --track origin/release-2010.12
Dec 03 09:01:48 <eFfeM_work> ok that seems to have doen the trick
Dec 03 09:02:05 <ericben> http://www.kernel.org/pub/software/scm/git/docs/git-checkout.html -> see explanation for --track
Dec 03 09:02:16 <eFfeM_work> thanks!
Dec 03 09:02:31 <ericben> "Furthermore, it directs git pull without arguments to pull from the upstream when the new branch is checked out. "
Dec 03 09:51:40 <Jay7> morning
Dec 03 10:00:26 <Jay7> http://tinderbox.openembedded.net/packages/1171052/
Dec 03 10:00:26 <bluelightning> hi Jay7
Dec 03 10:00:31 <Jay7> ffmpeg failed
Dec 03 10:00:49 <Jay7> bluelightning: hi again :)
Dec 03 10:01:57 <bluelightning> Jay7: I'm at work now, was at home before :)
Dec 03 10:02:46 <Jay7> bluelightning: I've thought you have deep night now (US)
Dec 03 10:03:11 <bluelightning> Jay7: nah, I live in the UK
Dec 03 10:05:31 <tasslehoff> back to basic. what is it that decides where BitBake will find my local.conf? I'm cleaning up / rewriting my env-setup.
Dec 03 10:06:53 <anarsoul> bluelightning: hi there
Dec 03 10:06:59 <bluelightning> hi anarsoul
Dec 03 10:07:28 <anarsoul> bluelightning: why opie-console does not like alsamixer? :)
Dec 03 10:13:31 <anarsoul> bluelightning: here's screenshot: http://fenix-fen.at.tut.by/opie-console.png
Dec 03 10:24:55 <hrw> anarsoul: you mean 'why is uses blocks for drawing'?
Dec 03 10:25:21 <anarsoul> hrw: yep
Dec 03 10:26:44 <hrw> try other font?
Dec 03 10:29:30 <bluelightning> anarsoul: hmm, could be a charset issue, or the font doesn't have the appropriate characters
Dec 03 10:30:23 <anarsoul> bluelightning: I see
Dec 03 10:32:16 <bluelightning> anarsoul: I'll test this myself when I get home
Dec 03 10:32:36 <bluelightning> anarsoul: btw good that you managed to fix the problem with qvfb :)
Dec 03 10:33:19 <bluelightning> anarsoul: btw are you subscribed to opie-devel@ ml?
Dec 03 10:33:28 <anarsoul> bluelightning: yep
Dec 03 10:33:35 <bluelightning> great
Dec 03 10:34:00 <anarsoul> bluelightning: I also got rid of SDL_main in libsdl-qpe, now just need to fix 32bpp support and clean up code a bit
Dec 03 10:36:45 <hrw> I thought that opie died already
Dec 03 10:37:55 <mckoan> to whom interested to FOSDEM, Ryanair special offers until Dec 6th
Dec 03 10:39:17 * hrw easyjet
Dec 03 10:42:04 <bluelightning> hrw: not yet ;)
Dec 03 10:44:40 <ka6sox> ericben, you still here?
Dec 03 10:45:19 <hrw> ka6sox: thx for nuke
Dec 03 10:45:23 <ericben> ka6sox: yes
Dec 03 10:45:42 <ka6sox> hrw, please pave when you are done :D
Dec 03 10:45:43 <Jay7> http://tinderbox.openembedded.net/packages/1172053/
Dec 03 10:45:46 <Jay7> libgpod is failed
Dec 03 10:47:34 <eFfeM_work> slugos build for release still failing:
Dec 03 10:47:53 <hrw> ka6sox: in meantime I would dig for massremoveusers plugins as recently spamers create new accounts and do one post
Dec 03 10:47:54 <Jay7> both (libgpod and ffmpeg) are failed on release-2010.12
Dec 03 10:48:10 <eFfeM_work> NOTE: package ipkg-utils-native-1.6+cvs20050404-r25: task do_fetch: Started NOTE: fetch http://anoncvs@sources.openembedded.org//ipkg-utils_anoncvs.handhelds.org__20060106.tar.gz --2010-12-03 10:23:42--  http://anoncvs@sources.openembedded.org//ipkg-utils_anoncvs.handhelds.org__20060106.tar.gz Resolving sources.openembedded.org... 140.211.169.165 Connecting to sources.openembedded.org|140.211.169.165|:80... connected. HTTP request sent, awai
Dec 03 10:48:30 <ka6sox> hrw, is that a search term or the name of the extension?
Dec 03 10:48:41 <Jay7> angstrom-2008.1/angstrom-gnome-image-big for collie and tosa
Dec 03 10:50:05 <hrw> ka6sox: search term
Dec 03 10:51:56 <hrw> ka6sox: http://www.mediawiki.org/wiki/Extension:DeleteBatch maybe?
Dec 03 10:55:05 <ka6sox> that one is like nuke
Dec 03 10:55:31 <ka6sox> I've found an external way to do it...looking to see if anyone has written this up more formally
Dec 03 10:56:36 <ka6sox> http://www.mediawiki.org/wiki/Extension_talk:User_Merge_and_Delete#How_to_merge_and_delete_a_large_list_of_users
Dec 03 10:57:16 <ka6sox> the basic idea is to merge all the bad ones into 1 account then delete that one account
Dec 03 10:59:18 <ka6sox> hrw, okay I'll look tommorrow....3am here.
Dec 03 10:59:46 <stefan_schmidt> ka6sox: go to bed!!
Dec 03 10:59:51 <stefan_schmidt> ka6sox: sleep well :)
Dec 03 10:59:58 <ka6sox> thanks, 
Dec 03 11:00:07 <ka6sox> stefan_schmidt, when is next patch merge?
Dec 03 11:00:36 <stefan_schmidt> I was thinking this we, but as the release is not out yet I will postpone it
Dec 03 11:00:45 <stefan_schmidt> ka6sox: You need more time for patchwork?
Dec 03 11:00:50 <ka6sox> yes, please
Dec 03 11:00:56 <stefan_schmidt> ka6sox: I would do either next weekend or the one after
Dec 03 11:00:59 <ka6sox> I will continue to hack on it.
Dec 03 11:01:02 <stefan_schmidt> ka6sox: thats fine
Dec 03 11:01:10 <stefan_schmidt> ka6sox: tell me when you are ready
Dec 03 11:01:21 <ka6sox> stefan_schmidt, will do
Dec 03 11:01:22 <ka6sox> nite
Dec 03 11:01:28 <stefan_schmidt> ka6sox: I appreciate your admin work and don't want to put you under pressure
Dec 03 11:01:42 <stefan_schmidt> it can wait as long as you need :)
Dec 03 11:01:43 <stefan_schmidt> night
Dec 03 11:03:05 <hrw> ka6sox: I know that it is late for you
Dec 03 11:27:50 <tasslehoff> Can anyone guess why this doesn't work: BBFILES ?= "${OE_SRCDIR}/recipes/*/*.bb", while it works if I replace ${OE_SRCDIR} with the output of echo $OE_SRCDIR?
Dec 03 11:28:20 <tasslehoff> which is /home/tasslehoff/src/openembedded
Dec 03 11:29:44 <CIA-101> 03Philip Balister <philip@balister.org> 07org.openembedded.dev * r4b6f049e48 10openembedded.git/recipes/uhd/ (3 files): 
Dec 03 11:29:44 <CIA-101> uhd-firmware : Convert to inc file and update recipe.
Dec 03 11:29:44 <CIA-101> Recipe can now easily be updates when new fpga images are released. Package
Dec 03 11:29:44 <CIA-101> is arch all, since it the fpga firmware is not machine dependent in general.
Dec 03 11:34:36 <tasslehoff> because stupid tasslehoff had forgotten to export the variable...
Dec 03 11:47:38 <woglinde> ping blindvt
Dec 03 12:00:56 <woglinde> hi obi
Dec 03 12:01:22 <obi> hi woglinde :)
Dec 03 12:02:13 <woglinde> obi still no rw access?
Dec 03 12:03:30 <obi> wow, it works for the first time today
Dec 03 12:03:43 <obi> yesterday it didn't
Dec 03 12:04:27 <woglinde> lol
Dec 03 12:05:08 <woglinde> hrms
Dec 03 12:05:10 <woglinde> | ./include/linux/socket.h:19:24: error: redefinition of typedef 'sa_family_t'
Dec 03 12:05:11 <woglinde> | ./include/bits/sockaddr.h:29:28: note: previous declaration of 'sa_family_t' was here
Dec 03 12:05:31 <woglinde> uclibc compile
Dec 03 12:07:24 <GNUtoo|laptop> hi woglinde 
Dec 03 12:07:30 <GNUtoo|laptop> hi GarthPS 
Dec 03 12:10:41 <GNUtoo|laptop> ERROR: Could not include required file openembedded/recipes/uhd/uhd-firmware.inc
Dec 03 12:11:18 <GNUtoo|laptop> I'll look
Dec 03 12:11:27 <GNUtoo|laptop> but openembedded/recipes/uhd/uhd-firmware.inc seem wrong
Dec 03 12:11:57 <GNUtoo|laptop> indeed
Dec 03 12:11:59 <GNUtoo|laptop> uhd-firmware_0001.20101124013843.905f426.bb:require openembedded/recipes/uhd/uhd-firmware.inc
Dec 03 12:12:35 <GNUtoo|laptop> hi pb__ 
Dec 03 12:13:09 <GNUtoo|laptop> pb__, 4b6f049e488b0e85f317d64b70af06920a1e2c47 broke parsing
Dec 03 12:15:22 <GNUtoo|laptop> oops sorry
Dec 03 12:15:50 <GNUtoo|laptop> I just realized that pb is... Phil Blundell
Dec 03 12:16:01 <GNUtoo|laptop> and not Philip Balister
Dec 03 12:16:23 <GNUtoo|laptop> what's Philip Balister 's nickname on irc?
Dec 03 12:16:43 <woglinde> gnutoo haha
Dec 03 12:16:58 <GNUtoo|laptop> indeed, I was not very awake
Dec 03 12:16:58 <woglinde> crfoton
Dec 03 12:16:59 <JaMa> Crofton|work: Crofton 
Dec 03 12:17:01 <GNUtoo|laptop> ok
Dec 03 12:17:04 <woglinde> ups crofton
Dec 03 12:17:11 <GNUtoo|laptop> Crofton, hi
Dec 03 12:17:47 <GNUtoo|laptop> Crofton, could you modify 4b6f049e488b0e85f317d64b70af06920a1e2c47 and change that: uhd-firmware_0001.20101124013843.905f426.bb:require openembedded/recipes/uhd/uhd-firmware.inc so it would parse correctly for me?
Dec 03 12:18:07 <woglinde> hm seems older kernel includes
Dec 03 12:18:37 <JaMa> GNUtoo|laptop: I think you can push fix for this without any ACK
Dec 03 12:20:10 <GNUtoo|laptop> ok
Dec 03 12:20:21 <GNUtoo|laptop> so I replace with require uhd-firmware.inc 
Dec 03 12:20:54 <GNUtoo|laptop> ?
Dec 03 12:21:09 <GNUtoo|laptop> basically I bet he must have been using layers or something like that
Dec 03 12:22:19 <JaMa> yes
Dec 03 12:22:30 <pb__> hi GNUtoo|laptop 
Dec 03 12:22:38 <pb__> oh, right, yeah, that is not me
Dec 03 12:22:46 <GNUtoo|laptop> pb__, sorry I wasn't very awake
Dec 03 12:28:41 <GNUtoo|laptop> I bet I don't need to bump PR
Dec 03 12:28:49 <GNUtoo|laptop> as it didn't parse and now it does
Dec 03 12:28:58 <JaMa> yes
Dec 03 12:28:58 <pb__> right
Dec 03 12:33:05 <Crofton> GNUtoo|laptop, what is he error?
Dec 03 12:33:24 <GNUtoo|laptop> Crofton, here:   ERROR: Could not include required file openembedded/recipes/uhd/uhd-firmware.inc
Dec 03 12:33:28 <GNUtoo|laptop> I'm preparing a fix
Dec 03 12:33:32 <Crofton> hmmm
Dec 03 12:33:44 <GNUtoo|laptop> maybe it worked for you 
Dec 03 12:33:44 <Crofton> this is a nuisance
Dec 03 12:33:47 <Crofton> yeah
Dec 03 12:33:50 <Crofton> I use layers
Dec 03 12:33:50 <GNUtoo|laptop> because your dir is openembedded
Dec 03 12:33:55 <Crofton> right
Dec 03 12:33:56 <GNUtoo|laptop> and mine is org.openembedded.dev
Dec 03 12:33:57 <Crofton> ah
Dec 03 12:34:01 <Crofton> yeah
Dec 03 12:34:09 <GNUtoo|laptop> so I fix it?
Dec 03 12:34:19 <Crofton> go ahed
Dec 03 12:34:25 <Crofton> I am in a hotel
Dec 03 12:34:32 <Crofton> you are going to drop the path?
Dec 03 12:34:50 <GNUtoo|laptop> -require openembedded/recipes/uhd/uhd-firmware.inc
Dec 03 12:34:50 <GNUtoo|laptop> +require uhd-firmware.inc
Dec 03 12:34:57 <GNUtoo|laptop> or should I do that instead:
Dec 03 12:35:05 <GNUtoo|laptop> require recipes/uhd/uhd-firmware.inc
Dec 03 12:35:07 <Crofton> go ahead
Dec 03 12:35:34 <GNUtoo|laptop> I'll do the frst solution as it's more natural
Dec 03 12:35:35 <Crofton> let me see if that works for me
Dec 03 12:36:24 <GNUtoo|laptop> ok
Dec 03 12:36:28 <GNUtoo|laptop> I wait for pushing then
Dec 03 12:36:31 <GNUtoo|laptop> for your testing
Dec 03 12:36:42 <Crofton> ok
Dec 03 12:36:46 <Crofton> testing now
Dec 03 12:37:06 <Crofton> when I came up with the path I had a typo and was watching a talk :)
Dec 03 12:37:15 <GNUtoo|laptop> ok lol
Dec 03 12:37:30 <Crofton> -require openembedded/recipes/uhd/uhd-firmware.inc
Dec 03 12:37:31 <Crofton> +require recipes/uhd/uhd-firmware.in
Dec 03 12:37:36 <Crofton> does this solve the problem for you?
Dec 03 12:37:47 <GNUtoo|laptop> yes of course it will, do you prefer that?
Dec 03 12:37:51 <Crofton> yes
Dec 03 12:37:55 <GNUtoo|laptop> ok
Dec 03 12:37:58 <GNUtoo|laptop> then I'll try it
Dec 03 12:37:58 <Crofton> it works better with layers
Dec 03 12:38:02 <GNUtoo|laptop> I know
Dec 03 12:38:07 <Crofton> try it. and I can push
Dec 03 12:38:11 <GNUtoo|laptop> ok
Dec 03 12:38:33 <Crofton> leave a note here, and Iwill commit before I go downstairs
Dec 03 12:38:58 <Crofton> need to pack up, goig home this afternoon
Dec 03 12:38:59 <GNUtoo|laptop> ok, else I can commit if you want
Dec 03 12:39:07 <GNUtoo|laptop> hmmm
Dec 03 12:39:33 <Crofton> I have it ready to go
Dec 03 12:39:35 <Crofton> thanks though
Dec 03 12:39:36 <GNUtoo|laptop> I made a typo
Dec 03 12:39:41 <GNUtoo|laptop> re-parsing
Dec 03 12:39:51 <Crofton> heh, I was doing that
Dec 03 12:40:10 <Crofton> I've run into a couple of people at the softare radio conf. who have used OE
Dec 03 12:40:14 <GNUtoo|laptop> in some sec I'll got the confirmation
Dec 03 12:40:17 <GNUtoo|laptop> it works
Dec 03 12:40:20 <Crofton> good to see us getting into lots of markets
Dec 03 12:40:22 <Crofton> ok
Dec 03 12:40:24 <Crofton> pushing
Dec 03 12:42:40 <CIA-101> 03Philip Balister <philip@balister.org> 07org.openembedded.dev * raab499a31d 10openembedded.git/recipes/uhd/uhd-firmware_0001.20101124013843.905f426.bb: uhd-firmware : fix bad path in require line.
Dec 03 12:45:23 <GNUtoo|laptop> works thanks a lot
Dec 03 12:47:03 <Crofton> np, thanks for catching that quickly
Dec 03 12:52:03 <risca> Hi! I have some trouble compiling libxcb for MACHINE="ea3250". The build fail with xproto.h not found. It seems it fails to include "usr/include/X11"
Dec 03 12:52:57 <risca> http://pastebin.com/6vG1htGy
Dec 03 12:53:26 <woglinde> risca build xproto before
Dec 03 12:53:42 <GNUtoo|laptop> #
Dec 03 12:53:42 <GNUtoo|laptop> | xcb.h:163:20: error: xproto.h: No such file or directory
Dec 03 12:53:48 <risca> Yeah, I forgot to mention that I've already built xproto
Dec 03 12:53:49 <woglinde> so it seems libxcb has nit enough deps
Dec 03 12:54:02 <GNUtoo|laptop> so first, where is xproto.h
Dec 03 12:54:03 <risca> and xproto.h is present in usr/include/X11
Dec 03 12:54:06 <GNUtoo|laptop> do you have it staged
Dec 03 12:54:07 <GNUtoo|laptop> ok
Dec 03 12:54:22 <GNUtoo|laptop> usr/include/x11 is in your tmpdir I guess
Dec 03 12:54:49 <risca> Present in tmp/sysroots/armv5te-angstrom-linux-gnueabi/usr/include/X11/
Dec 03 12:54:51 <GNUtoo|laptop> like tmp/sysroot/armv-something/usr/include/X11/xproto.h
Dec 03 12:54:52 <GNUtoo|laptop> ok
Dec 03 12:55:02 <risca> That's right
Dec 03 12:55:19 <GNUtoo|laptop> so does ./configure --help shows something about X11 dir?
Dec 03 12:55:31 <risca> Wait, I'll check
Dec 03 12:55:53 <risca> No
Dec 03 12:56:12 <GNUtoo|laptop> could you pastebin ./configure --help ?
Dec 03 12:56:16 <risca> Shure
Dec 03 12:57:03 <risca> http://pastebin.com/g5ff6ysv
Dec 03 12:58:44 <risca> I had som errors before about fop and had to add EXTRA_OECONF+="--with-fop=no" to several recipes (X11 related)
Dec 03 12:58:51 <risca> Don't know if that is relevant
Dec 03 12:59:15 <woglinde> dont think so
Dec 03 12:59:26 <risca> No, neither do I
Dec 03 12:59:37 <GNUtoo|laptop> woglinde, what should we advise now CFLAGS += " -I${STAGING_INCDIR}/X11 " ?
Dec 03 13:00:13 <JaMa> risca: please send patch with those --with-fop=no
Dec 03 13:00:36 <woglinde> xcb builds here fine
Dec 03 13:00:39 <JaMa> risca: it detects fop from host and it shouldn't
Dec 03 13:00:53 <risca> I see
Dec 03 13:04:34 <risca> Should I add CFLAGS += " -I${STAGING_INCDIR}/X11 " to libxcb.inc ?
Dec 03 13:07:58 <Jay7> do we have any fixes in master for ffmpeg_svn?
Dec 03 13:08:18 <Jay7> it fails to build in release branch 
Dec 03 13:08:33 <woglinde> jay7 at which point?
Dec 03 13:08:39 <Jay7> http://tinderbox.openembedded.net/packages/1174039/
Dec 03 13:09:21 <Jay7> ah.. I've remember something about submodules
Dec 03 13:13:01 <risca> JaMa, Should I send the patch to oe dev mailing list?
Dec 03 13:14:26 <JaMa> risca: yes, please do
Dec 03 13:15:56 <risca> It should be sent now
Dec 03 13:15:57 <risca> I think
Dec 03 13:16:07 <risca> Kind of new to git send-email
Dec 03 13:26:34 <risca> Strange... Remember how I said I had xproto.h in 'usr/include/X11'. Well, it turns out I had Xproto.h in there (notice the capital X)
Dec 03 13:31:39 <obi> Jay7: the fix is to upload the snapshot tarball to your sources mirror
Dec 03 13:32:06 <Jay7> obi: it's not my personal problem :)
Dec 03 13:32:21 <Jay7> I've even not used angstrom-gnome-image[-big] :)
Dec 03 13:33:19 <Jay7> we should do something with this recipe or upload tarball to oe's mirror
Dec 03 13:34:32 <obi> Jay7: well, that's what i meant to say
Dec 03 13:34:47 <Jay7> I'll post this to ML
Dec 03 13:34:56 <obi> the file is available on the angstrom mirror though
Dec 03 13:35:16 <Jay7> not a good thing for release..
Dec 03 13:35:28 <Jay7> hm..
Dec 03 13:35:51 <Jay7> shouldn't angstrom building automatically use angstrom mirrors?
Dec 03 13:36:05 <woglinde> yes
Dec 03 13:36:08 <obi> it did for me
Dec 03 13:36:11 <woglinde> tries first normal
Dec 03 13:36:18 <woglinde> and than swtich to mirros
Dec 03 13:36:21 <Jay7> why is it broken then?
Dec 03 13:36:39 <obi> but i disabled the mirror, because it was terribly slow. that's how i came to see this ffmpeg build error in the first place ;)
Dec 03 13:37:14 <obi> maybe the mirror was down when you fetched the sources
Dec 03 13:38:17 <obi> i have a recipe for ffmpeg 0.6.1, but it's almost untested
Dec 03 13:38:36 <woglinde> hm did they release?
Dec 03 13:38:59 <obi> yes
Dec 03 13:39:05 <woglinde> oha
Dec 03 13:39:55 <obi> http://git.opendreambox.org/?p=obi/openembedded.git;a=commitdiff;h=183099bf2e25ac724489bd964773cd7e1e3d9d6c
Dec 03 13:40:38 <obi> the md5sum and shasum are missing from the recipe, because i've only built it with an old version of OE
Dec 03 13:41:24 <obi> besides that, i guess the recipe is ok. i disabled the gpl and non-free code however
Dec 03 13:46:44 <risca> I'm still having problems with xproto.h
Dec 03 13:47:14 <risca> I did a symlink to Xproto.h for a quick fix, but xproto.h wasn't the only file missing.
Dec 03 13:47:43 <woglinde> risca as I said you are missing some xproto stuff
Dec 03 13:47:59 <risca> But what more than bitbake xproto do I have to do?
Dec 03 13:48:14 <woglinde> hm hm
Dec 03 13:48:21 <woglinde> intresstin xproto.h is in libxcb-dev
Dec 03 13:48:22 <woglinde> hm
Dec 03 13:48:31 <risca> Hmm... I see...
Dec 03 13:48:50 <risca> One of the other missing files (bigreq.h) is in there as well
Dec 03 13:49:09 <woglinde> but maybee I havent installed all xproto packages here
Dec 03 13:49:12 <woglinde> hm
Dec 03 13:49:53 <Jay7> libgpod failure: http://tinderbox.openembedded.net/packages/1172053/
Dec 03 13:49:57 <Jay7> looks like race
Dec 03 13:50:12 <Jay7> single failure only for tosa build
Dec 03 13:50:21 <hosuk> hello~
Dec 03 13:52:05 <risca> woglinde, Since there is no libxcb-dev recipe in OE, do I have to create one?
Dec 03 13:52:11 <woglinde> no no
Dec 03 13:52:36 <risca> Yeah, I remember. xcb compiles on your system
Dec 03 13:53:44 <woglinde> hm
Dec 03 13:54:00 <risca> Heavy thinking goin' on, I guess
Dec 03 13:54:06 <woglinde> no
Dec 03 13:54:12 <woglinde> Xproto.h should be enough
Dec 03 13:54:18 <woglinde> for compiling libxcb
Dec 03 13:54:39 <risca> I tried doing a ln -sv Xproto.h xproto.h
Dec 03 13:54:47 <woglinde> no
Dec 03 13:55:01 <woglinde> Xproto.h are the protoheaders
Dec 03 13:55:09 <woglinde> xproto.h is from libxcb self
Dec 03 13:55:13 <risca> Ah, I see
Dec 03 13:55:19 <risca> Tricky
Dec 03 13:56:32 <risca> Then, how come libxcb fails to see its own files?
Dec 03 13:57:02 <woglinde> hm could you provide the config.log from configure?
Dec 03 13:57:07 <risca> shure
Dec 03 14:00:42 <risca_> My PC froze! Had to do alt-sysrq-b :S
Dec 03 14:03:10 <risca> http://pastebin.com/6QSpYzYa <-- config.log
Dec 03 14:03:19 <cpo> hi, im wondering if thers an alternative to use ifup on openembedded for starting network during boot progress?
Dec 03 14:03:33 <woglinde> cpo networkmanager
Dec 03 14:03:36 <woglinde> or conman
Dec 03 14:03:46 <cpo> because ifup reads dhcp hostname from the interfaces file instead of /etc/hostname
Dec 03 14:04:03 <woglinde> ?
Dec 03 14:04:29 <cpo> i have to provide the dhcp hostname i wanna use for my interface in /etc/network/interfaces. but i want to use the one from /etc/hostname
Dec 03 14:04:36 <cpo> isn't that the intention of /etc/hostname?
Dec 03 14:04:42 <cpo> or, a bit of its intention
Dec 03 14:05:20 <cpo> or did i miss the point?
Dec 03 14:05:53 <woglinde> risca hm dont see any error in this
Dec 03 14:07:19 <risca> Is xproto (the recipe) related to libxcb? Could the error originate from a faulty xproto build?
Dec 03 14:07:49 <woglinde> not really
Dec 03 14:07:50 <cpo> risca#: bitbake -g <recipe> to get the dependencies
Dec 03 14:07:59 <risca> cpo, thanks
Dec 03 14:08:46 <woglinde> hm
Dec 03 14:08:51 <woglinde> but we have this patch
Dec 03 14:09:01 <woglinde> ./xcb/libxcb/configure_git.patch
Dec 03 14:09:11 <woglinde> -XCBPROTO_XCBINCLUDEDIR=`$PKG_CONFIG --variable=xcbincludedir xcb-proto`
Dec 03 14:09:11 <woglinde> +XCBPROTO_XCBINCLUDEDIR=${STAGING_DIR}`$PKG_CONFIG --variable=xcbincludedir xcb-proto`
Dec 03 14:09:25 <woglinde> that might be the problem
Dec 03 14:10:30 <risca> You mean there is a faulty patch, or the patch is not applied?
Dec 03 14:10:51 * Jay7 afk
Dec 03 14:10:55 <woglinde> no no the patch is applied
Dec 03 14:11:06 <woglinde> but maybee some strange setup at your side
Dec 03 14:11:20 <risca> Perhaps
Dec 03 14:11:39 <risca> I'm running Sabayon linux 5.0 (gentoo-based)
Dec 03 14:11:43 <woglinde> which branch are you using?
Dec 03 14:11:59 <obi> kergoth: i'm currently trying to convert our distro to using bblayers. one thing i noticed is, that unlike without layers, i cannot run bitbake from anywhere i like (given that the environment was set up correctly), because bitbake looks for conf/bblayers.conf in $PWD and its parents, if i understand the code correctly. how about adding an environment variable to override the location of bblayers.conf?
Dec 03 14:12:19 <risca> woglinde, Which branch of what?
Dec 03 14:12:45 <woglinde> risca oe
Dec 03 14:13:15 <risca> The one mentioned in the wiki (getting started)
Dec 03 14:13:16 <risca> git://git.openembedded.org/openembedded
Dec 03 14:13:17 <woglinde> re crofton
Dec 03 14:13:33 <woglinde> so its master
Dec 03 14:13:41 <risca> Yeah
Dec 03 14:15:44 <woglinde> hm as I said no problems here compiling libxcb
Dec 03 14:16:02 <risca> I'm thinking of firing up a Virtual PC for OE development. Any suggestions for distro?
Dec 03 14:16:45 <woglinde> no
Dec 03 14:17:56 <obi> with bblayers, is there a nicer way than defining THISDIR and modifiying FILESPATH manually in every .bbappend file to add a directory to the search path?
Dec 03 14:20:41 <obi> poky has FILESEXTRAPATHS, which uses a predefined THISDIR, which looks a little better. but i don't understand bitbake well enough to implement that. and i don't want to duplicate work, in case there already is a different solution ;)
Dec 03 14:21:03 <bluelightning> ka6sox: btw, I have split archives for opie ready to upload this weekend
Dec 03 14:21:41 <woglinde> obi discuss it with kergoth
Dec 03 14:22:21 <obi> ok
Dec 03 14:23:44 <Crofton> hi
Dec 03 14:23:45 <foerster> obi: I use amend.inc instead of bbappend, with FILESPATHBASE set in local.conf to contain my overlay all the time.  
Dec 03 14:27:04 <obi> i thought about that, too. i want to use 4 layers on top of OE. can FILESPATHBASE also be set in each layer's layer.conf?
Dec 03 14:29:26 <foerster> obi: not sure, I don't actually use bblayers or layer.conf
Dec 03 14:30:11 <foerster> in local.conf, I just have the path to each layer in a variable, then FILESPATHBASE is set to contain those.
Dec 03 14:31:19 <obi> ok, i'll try that
Dec 03 14:31:50 <foerster> obi: http://pastie.org/1344503
Dec 03 14:33:58 <obi> thanks
Dec 03 14:34:04 <foerster> np
Dec 03 15:07:11 <risca> woglinde, I think I found the problem with xproto.h and friends. Those files seem to be generated by python scripts. A quick check reveals that 'sysroots/i686-linux/usr/bin/python' has a filesize of 0kb
Dec 03 15:07:35 <risca> Something went wrong somewhere when building python
Dec 03 15:07:44 <risca> I'm doing this all over again
Dec 03 15:08:17 <woglinde> risca hm oh
Dec 03 15:08:18 <woglinde> yes
Dec 03 15:25:45 <woglinde> ping pb
Dec 03 15:28:25 <pb__> hi woglinde 
Dec 03 15:28:46 <woglinde> pb we have a problem with openjdk and newer libtool
Dec 03 15:29:25 <woglinde> one lib has zlib as additional linking lib
Dec 03 15:29:44 <woglinde> and libtool uses the .la from staging
Dec 03 15:29:56 <woglinde> for zlin
Dec 03 15:29:58 <woglinde> args zlib
Dec 03 15:30:11 <woglinde> in the .la file /usr/lib is given as filepath
Dec 03 15:30:24 <woglinde> so -lz becomes /usr/lib/libz.so
Dec 03 15:30:46 <woglinde> which errors crosscompiling
Dec 03 15:31:38 <pb__> hm, right.  I thought package-staging.bbclass was meant to mangle the .la files appropriately.
Dec 03 15:31:46 <woglinde> http://pastebin.com/JqZWgMaN
Dec 03 15:33:02 <woglinde> hm shoud we remove the .la files at all?
Dec 03 15:35:05 <woglinde> but I wonder whats wrong in this case
Dec 03 15:37:07 <woglinde> hm ah
Dec 03 15:37:18 <woglinde> yeah the linkerscript
Dec 03 15:37:39 <woglinde> hm hm
Dec 03 15:38:18 <woglinde> ah
Dec 03 15:39:24 <woglinde> right openjdk dont reruns autogen.sh
Dec 03 15:39:32 <woglinde> on ncacao
Dec 03 15:46:10 <eFfeM_work> JaMa: might have the missing ipkg files, can check later, need to go now
Dec 03 15:46:54 <JaMa> eFfeM_work: ?
Dec 03 15:47:41 <JaMa> eFfeM_work: you mean that you have them in another downloads dir and so can push them to sources mirror?
Dec 03 15:48:20 <woglinde> re stefan
Dec 03 15:48:35 <eFfeM_work> i think I have them, did not include our mirror on purpose to see if things build from scratch, but need to go now
Dec 03 15:48:49 <eFfeM_work> can't push myself will ask once I have found the file
Dec 03 15:49:00 <stefan_schmidt> hi woglinde 
Dec 03 15:49:04 <eFfeM_work> need to run, later (= tomorrow, won't be online tonight)
Dec 03 15:49:05 <JaMa> woglinde: with newer libtool you mean libtool with sysroot support enabled right?
Dec 03 15:49:06 <eFfeM_work> cya
Dec 03 15:49:24 <JaMa> woglinde: then there should be =/usr/lib in .la IIRC
Dec 03 15:50:11 <woglinde> jama I found the problem
Dec 03 15:50:19 <JaMa> woglinde: and = is replaced by sysroot value in libtool while building ie openjdk
Dec 03 15:50:31 <woglinde> as I said cacao 0.99.4 when building inside openjdk has older libtool stuff
Dec 03 15:50:47 <woglinde> so it dont replaces -lz with the right lib
Dec 03 15:51:21 <JaMa> ok
Dec 03 15:51:33 <woglinde> and yes libtool 2.4
Dec 03 15:51:55 <stefan_schmidt> ah, woglinde is fixing openjdk build with libtool 2.4 :)
Dec 03 15:52:16 <woglinde> I am now thinking how to run autoreconf from the openjdk Makefile
Dec 03 15:53:47 <stefan_schmidt> woglinde: No autgen.sh script you could sneak this in?
Dec 03 15:54:05 <woglinde> stefan not really
Dec 03 15:54:11 <woglinde> think about it
Dec 03 15:54:38 <woglinde> we might have to run the same stuff autotools.bbclass runs
Dec 03 15:56:29 <woglinde> maybee running libtoolize is enough
Dec 03 15:56:54 <woglinde> will try it
Dec 03 16:18:08 <woglinde> hms no
Dec 03 16:23:39 <beerpages> Erstelle dir deine eigene Umfrage - Beispiel: http://www.beerpages.de/view/2H/Anti-%22Die%20%C3%84rzte%22%20Kampagne
Dec 03 17:25:28 <CIA-101> 03Chris Larson <chris_larson@mentor.com> 07master * r57402ee727 10bitbake.git/bin/bitbake: 
Dec 03 17:25:28 <CIA-101> Error more pleasantly when trying to use python2.4
Dec 03 17:25:28 <CIA-101> Signed-off-by: Chris Larson <chris_larson@mentor.com>
Dec 03 17:48:28 <CIA-101> 03Chris Larson <chris_larson@mentor.com> 07master * r424428a764 10bitbake.git/lib/bb/cooker.py: 
Dec 03 17:48:28 <CIA-101> Fix the <100 recipe progress fix
Dec 03 17:48:28 <CIA-101> Signed-off-by: Chris Larson <chris_larson@mentor.com>
Dec 03 18:06:38 <morphis> heyho
Dec 03 18:53:23 * kergoth_ wonders if we should try to retain history when pulling the poky bitbake runqueue bits over, or just pull the code directly
Dec 03 18:53:26 * kergoth_ ponders
Dec 03 19:05:41 <B_Lizzard> Is bitbake shitting me?
Dec 03 19:05:48 <jconnolly> probably =/
Dec 03 19:06:01 <B_Lizzard> Uh huh
Dec 03 19:10:49 <B_Lizzard> When building my image, bluez-utils gets pulled in, which fails to compile
Dec 03 19:11:00 <B_Lizzard> bitbake bluez-utils builds bluez4
Dec 03 19:11:23 <B_Lizzard> And I cannot find which package pulls in bluez-whatever via bitbake -g
Dec 03 19:15:54 <B_Lizzard> Ah, task-base
Dec 03 19:17:54 <CIA-101> 03Eric Bénard <eric@eukrea.com> 07org.openembedded.dev * rb1c7bc4305 10openembedded.git/recipes/busybox/ (busybox-1.18.0/busybox-1.18.0-sha.patch busybox_1.18.0.bb): 
Dec 03 19:17:55 <CIA-101> busybox-1.18.0: add latest fix
Dec 03 19:17:55 <CIA-101> libbb/pw_encrypt_sha: forgot to move &ctx to 1st param here in sha hash rework
Dec 03 19:17:55 <CIA-101> http://git.busybox.net/busybox/commit/?id=5dcc6606cb6ef41e0e0b0f828f670cc61311a3f7
Dec 03 19:17:55 <CIA-101> Signed-off-by: Eric Bénard <eric@eukrea.com>
Dec 03 19:32:41 <B_Lizzard> Now it works
Dec 03 19:32:44 <B_Lizzard> Via magic, I presume
Dec 03 19:32:50 * B_Lizzard shoots himself
Dec 03 19:53:33 <khem> Hi all
Dec 03 19:53:47 <ka6sox> hi khem 
Dec 03 19:53:50 <khem> has everyone update the table of builds 
Dec 03 19:53:53 <khem> to release ?
Dec 03 19:53:56 <khem> hey ka6sox 
Dec 03 20:30:08 <gregoiregentil> Any idea how  I could get iconv (as a standalone executable) on my target? I'm using glibc and it provides virtual/libiconv. I'm confused.
Dec 03 20:31:16 <Jay7> evening
Dec 03 20:32:03 <Jay7> no good with angstrom-2010.x and angstrom-gnome-image-big too
Dec 03 20:32:20 <Jay7> now samba fails
Dec 03 20:33:01 <Jay7> and build for efika eats 32Gb of space on about 60% of build
Dec 03 20:34:10 <CIA-101> 03Khem Raj <raj.khem@gmail.com> 07master * r5d15f2d0f1 10openembedded.git/recipes/eglibc/ (eglibc-package.bbclass eglibc.inc): 
Dec 03 20:34:10 <CIA-101> eglibc: Move eglibc-static after eglibc-dev in order
Dec 03 20:34:10 <CIA-101> With the current order *nonshared.a ends up in eglibc-static
Dec 03 20:34:10 <CIA-101> which is not intended. The reorder will ensure that these files
Dec 03 20:34:10 <CIA-101> go into -dev
Dec 03 20:34:25 <CIA-101> 03Graham Gower <graham.gower@gmail.com> 07master * re967f4332e 10openembedded.git/recipes/bind/ (4 files): (log message trimmed)
Dec 03 20:34:25 <CIA-101> bind: Fix for DISTRO=micro: s|/usr/bin|${bindir}|
Dec 03 20:34:25 <CIA-101> This is a conservative change to fix micro. Bind needs more attention
Dec 03 20:34:25 <CIA-101> as the latest in tree version is from 2008 and several vulnerabilities
Dec 03 20:34:25 <CIA-101> have been found and fixed since then. If you are doing anything
Dec 03 20:34:30 <Jay7> http://tinderbox.openembedded.net/builders/Jay7-tb/
Dec 03 20:34:33 <CIA-101> 03Graham Gower <graham.gower@gmail.com> 07master * rf60820472e 10openembedded.git/recipes/openjade/openjade_1.3.2.bb: 
Dec 03 20:34:33 <CIA-101> openjade_1.3.2.bb: Create directory before installing a file to that location.
Dec 03 20:34:33 <CIA-101> Fixes a breakage I introduced in last openjade commit.
Dec 03 20:34:33 <CIA-101> Signed-off-by: Graham Gower <graham.gower@gmail.com>
Dec 03 20:34:33 <CIA-101> Signed-off-by: Khem Raj <raj.khem@gmail.com>
Dec 03 20:35:03 <CIA-101> is recommended. If you have several perl versions
Dec 03 20:35:03 <CIA-101> installed, select the one Automake should use using
Dec 03 20:35:42 <kergoth_> cia seems broken for pushes of large numbers of commits, it seems to drop chunks of the output but keep other chunks
Dec 03 20:35:55 <kergoth_> wonder if thats just the irc notification, or something more fundamental
Dec 03 20:37:00 <ka6sox> the hook in git I've seen problems with occasionally where it does that with CIA on various projects
Dec 03 20:41:07 <khem> Jay7: samba has issues on ARM/gcc-4.5
Dec 03 20:41:33 <khem> gregoiregentil: virtual/libiconv is a virtual provider it does not replace real binaries
Dec 03 20:41:43 <khem> its for resolving dependencies in metadata
Dec 03 20:41:51 <gregoiregentil> khem: thanks. but I'm still confused
Dec 03 20:42:04 <gregoiregentil> khem: so if I want to have an executable iconv
Dec 03 20:42:10 <gregoiregentil> on my target, what should I do?
Dec 03 20:42:13 <khem> gregoiregentil: real the documentation whatever little is out there 
Dec 03 20:43:07 <khem> gregoiregentil: iconv should be provided via a package from glibc/eglibc if you are using that
Dec 03 20:43:16 <khem> gregoiregentil: lemme see what package
Dec 03 20:43:54 <gregoiregentil> yes. I'm recompiling glibc without INHERIT += "rm_work to investigate more
Dec 03 20:44:03 <gregoiregentil> in the meantime, I have listed all glibc package
Dec 03 20:44:12 <gregoiregentil> but I haven't found any trace of iconv (as an exe)
Dec 03 20:45:24 <khem> hmm
Dec 03 20:46:36 <gregoiregentil> I think that I will have to patch glibc so that it creates this executable because I don't see it anywhere
Dec 03 20:46:43 <gregoiregentil> I'm surprised that nobody asked this question before...
Dec 03 20:47:19 <khem> its possible that we dont build it
Dec 03 20:47:28 <khem> because we generate those during build
Dec 03 20:48:02 <gregoiregentil> but in the recipe, I haven't found something like "--without-iconv"
Dec 03 20:48:26 <gregoiregentil> or should it be more something like EXTRA_OECONF += "--with-iconv" or whatever...
Dec 03 20:48:29 <gregoiregentil> ?
Dec 03 21:11:06 <khem> gregoiregentil: it needs to be compiled in first
Dec 03 21:11:11 <khem> make sure that its generated
Dec 03 21:11:31 <khem> then see if it is packaged into some package
Dec 03 21:11:39 <khem> generated by glibc recipe
Dec 03 21:14:39 <gregoiregentil> yup
Dec 03 21:14:53 <gregoiregentil> exactly. I haven't seen it before because it was rm -rf for the work directory
Dec 03 21:15:04 <gregoiregentil> cool. I may modify the recipe to add a glibc-iconv....ipk
Dec 03 21:15:13 <woglinde> good nite
Dec 03 21:31:03 <Jay7> hm.. do poky/yocto have runtime testing system?
Dec 03 21:31:17 <Jay7> something qemu-based
Dec 03 21:33:05 <kergoth> Jay7: they do automated runtime testing as a part of their autobuilder, yes
Dec 03 21:33:10 <kergoth> don't know how it's implemented though
Dec 03 21:33:15 <Jay7> hm..
Dec 03 21:40:01 <ka6sox> hiya sakoman 
Dec 03 21:40:24 <sakoman> hey ka6sox!
Dec 03 21:41:52 <khem> sakoman: hi
Dec 03 21:42:15 <khem> kergoth: safely back home ?
Dec 03 21:42:32 <ka6sox> I'd like to setup some kind of build-server characterizing either 1 time/ environment or maybe every time to determine what class of CPU/How much RAM/how fast the disks are and create a metric that can be used by bitbake to determine how to schedule the build and maximize performance based on hardware resource.
Dec 03 21:42:37 <sakoman> khem: yes, made it back around 5pm yesterday
Dec 03 21:42:45 <CIA-101> 03Koen Kooi <koen@openembedded.org> 07org.openembedded.dev * reffc276f92 10openembedded.git/recipes/gcc/ (gcc-4.3.3.inc gcc-4.5.inc): 
Dec 03 21:42:45 <CIA-101> gcc 4.3.3, 4.5: bump PR for gcc-package-target.inc fixes
Dec 03 21:42:45 <CIA-101> Signed-off-by: Koen Kooi <koen@openembedded.org>
Dec 03 21:46:23 <Jay7> btw, good idea to setup-testbuilder
Dec 03 21:53:13 <khem> kergoth: ok so I created the final release tag locally now if I do git push --tags it will push it but this tag is on a branch right now later I will merge the branch back into master will git push --tags rely on branch information
Dec 03 21:53:47 <khem> kergoth: iow I wont be killing some information that tag needs by merging and then deleting the branch right ?
Dec 03 21:54:47 <khem> actually I just found a way to keep the tag moving without branching next release cycle we will use that approach 
Dec 03 21:55:19 <khem> without creating a branch but having a release tag which keeps moving as we sanitize the release
Dec 03 21:55:43 <kergoth> khem: the tag just points at a commit.
Dec 03 21:55:54 <kergoth> it doesn't care whether a branch exists that also points at that commit
Dec 03 21:55:56 <khem> kergoth: yeah thats what I thought
Dec 03 21:56:06 <khem> ok
Dec 03 21:56:23 <khem> I am waiting for the final build to finish
Dec 03 21:56:26 <khem> and then its up
Dec 03 21:56:50 <khem> kergoth: I wonder if I will be able to delete the branch though ?
Dec 03 21:56:55 <kergoth> of course
Dec 03 21:57:00 <kergoth> again, the tag doesn't carea bout teh commit
Dec 03 21:57:02 <kergoth> er about the branch
Dec 03 21:57:03 <khem> I mean the remote branch
Dec 03 21:57:19 <kergoth> oh, might ahve to ask cbrake or someone to drop it
Dec 03 21:57:24 <khem> ook
Dec 03 21:57:52 <khem> there are no commits that are there on the branch which are not in master does it make much sense to merge it into master ?
Dec 03 21:58:14 <khem> it will only create one commit saying that branch is merged into master I think thats it
Dec 03 22:00:30 <kergoth> right, that's it.  it marks it, basically.  this makes it so a log comparing master against the tag won't show the user any changes which exist on the tag but not in master
Dec 03 22:00:34 <kergoth> which i think is worthwhile
Dec 03 22:06:42 <khem> ok right
Dec 03 22:12:31 <ka6sox> let me know if we need to drop it
Dec 03 22:12:45 <khem> ka6sox: you mean the branch ?
Dec 03 22:13:13 <ka6sox> yes
Dec 03 22:25:12 <Jay7> are we released finally? :)
Dec 03 22:28:29 <khem> Jay7: few more moments
Dec 03 22:28:59 <Jay7> just don't forget to announce in ML's and on site :)
Dec 03 22:29:06 <khem> yes sure
Dec 03 22:36:33 <khem> hmm I wonder if we can create tar files from our git server
Dec 03 22:44:02 <coreyfro> Hey all.  When I execute ' MACHINE="overo" bitbake base-image ' I get the error "x-load_git.bb do_deploy failed" but when I execute ' MACHINE="beagleboard" bitbake base-image ' I sucessfully build a filesystem.  I am new to OE (being a gentoo nerd) so I don't know how to sift out a solution, but seeing as how similar these platforms are and I succeded in building one, there must be a problem with how the overo is defined, but I don't k
Dec 03 22:47:36 <khem> coreyfro: beagleboard pins the recipe to different git rev
Dec 03 22:47:54 <khem> so even though its same recipe it builds differently for overo and beagle
Dec 03 22:48:08 <khem> coreyfro: what error do u get exactly
Dec 03 23:00:04 <coreyfro> KHEM: one second
Dec 03 23:00:29 <coreyfro> KHEM: NOTE: Task failed: ('function do_deploy failed', '/home/coreyfro/angstrom-setup-scripts/build/tmp-angstrom_2008_1/work/overo-angstrom-linux-gnueabi/x-load-1_1.42+r17+gitr319b26586fafb86f919f514bcd175838aaab96b3-r17/temp/log.do_deploy.31023')
Dec 03 23:00:38 <coreyfro> catting that log gives me:
Dec 03 23:00:56 <coreyfro> install: cannot stat `/home/coreyfro/angstrom-setup-scripts/build/tmp-angstrom_2008_1/work/overo-angstrom-linux-gnueabi/x-load-1_1.42+r17+gitr319b26586fafb86f919f514bcd175838aaab96b3-r17/git/x-load.bin.ift': No such file or directory
Dec 03 23:00:56 <coreyfro> ERROR: Function do_deploy failed
Dec 03 23:04:18 <florian> re
Dec 03 23:04:51 <florian> Is there a way to tell bitbake to return non-zero if there is a critical error only?
Dec 03 23:05:21 <Jay7> imho no, but better to ask kergoth :)
Dec 03 23:06:27 <florian> no translates to "vi" ;)
Dec 03 23:06:50 <Jay7> $EDITOR ;)
Dec 03 23:07:13 <khem> coreyfro: does x-load.bin.ift exist in the build tree somewhere
Dec 03 23:07:14 <Jay7> I'm really want this too
Dec 03 23:07:41 <Jay7> this will allow to know what was build status w/o looking into logs/oestats
Dec 03 23:11:05 <coreyfro> khem: there isn't
Dec 03 23:13:42 <coreyfro> khem: I'm trying to figure out what x-load is (other than "x-load for TI OMAP3 based systems"), if I need it, and what is calling it as a dependencie.  If I don't need it, I'd just as soon get rid of it.
Dec 03 23:14:03 <coreyfro> khem: if it's some w windows dependency, I don't need it
Dec 03 23:16:42 <coreyfro> khem: I am looking at the dot dependency tree output.  I don't see what's calling it
Dec 03 23:18:38 <coreyfro> console-image pulls it, too
Dec 03 23:20:49 <khem> coreyfro: well its some sort of tool needed for NAND flash I guess
Dec 03 23:39:47 <coreyfro> khem: OK, I think I need to kill it somehow.  I'll have to read up on developing my own images I guess.
Dec 03 23:39:53 <khem> coreyfro: ok conf/machine/include/omap3.inc:EXTRA_IMAGEDEPENDS += "u-boot x-load"
Dec 03 23:40:03 <khem> in that file get rid of x-load
Dec 03 23:40:15 <khem> on that given line
Dec 03 23:41:13 <khem> kergoth: http://www.linuxfordevices.com/c/a/News/Mentor-acquires-key-assets-of-CodeSourcery/ 
Dec 03 23:41:24 <khem> kergoth: you did not tell me about that :)
Dec 03 23:41:32 <khem> its interesting
Dec 03 23:42:06 <coreyfro> khem: Thanks! 
Dec 03 23:53:00 <Jay7> hm...
Dec 03 23:53:15 <Jay7> poky is using cross-localedef instead of qemu to generate locales
Dec 03 23:53:35 <khem> I know
Dec 03 23:53:45 <Jay7> good idea to reuse that :)
Dec 03 23:53:46 <khem> we should backport that to OE 
Dec 03 23:53:51 <khem> eglibc
Dec 03 23:53:58 <Jay7> yep
Dec 03 23:54:06 <khem> yes long ago I did that in MV's toolchain
Dec 03 23:54:30 <khem> Jay7: do u have hold of changes ?
Dec 03 23:54:35 <khem> would mind to backport 
Dec 03 23:55:53 <Jay7> not sure what do you mean under 'hold of changes' but I'm looking into http://git.pokylinux.org/cgit/cgit.cgi/poky/tree/meta/classes/libc-package.bbclass
Dec 03 23:56:06 <Jay7> shouldn't be hard to backport
Dec 03 23:56:22 <khem> Jay7: thats what I mean
Dec 03 23:57:11 <Jay7> better to do when you know python..
Dec 03 23:57:40 <khem> Jay7: I will do it after the release
Dec 03 23:58:07 <Jay7> khem: awesome! :)
Dec 03 23:59:14 <Jay7> there is runtime testing code around
Dec 03 23:59:22 <Jay7> simple enough
Dec 03 23:59:40 <Jay7> but I'm unsure that this is bitbake's task
Dec 04 00:03:17 <coreyfro> khem: cool, looks like I will not be building x-load now.
Dec 04 00:07:30 * Jay7 -> sleep
Dec 04 00:09:12 <ka6sox> nn Jay7 
Dec 04 00:09:40 <coreyfro> khem: WOOOOOOOOOOT! Thanks!
Dec 04 00:09:58 * coreyfro kicks x-load
Dec 04 00:20:59 <ka6sox> mwester, got a minute to help with a problem?
Dec 04 00:21:24 <ka6sox> oh sorry...Hi mwester!
Dec 04 00:23:00 <khem> who can edit the front page of wiki
Dec 04 00:23:06 <khem> especially news section ?
Dec 04 00:23:14 <khem> I dont have permissions to change that
Dec 04 00:24:34 <ka6sox> what do you need?
Dec 04 00:24:44 <ka6sox> an announcement on the release?
Dec 04 00:30:14 * mwester returns
Dec 04 00:30:30 <mwester> Greetings mr. ka6sox.  What's up?
Dec 04 00:35:20 <ka6sox> hiya mwester
Dec 04 00:35:46 <ka6sox> I need the source tgz for ipkg-native that slugOS uses
Dec 04 00:36:11 <ka6sox> the its currently in teh bb as coming from handhelds.org (which has been down for a LONG time)
Dec 04 00:36:24 <ka6sox> I want to put it in the OE sourcemirror.
Dec 04 00:37:50 <mwester> Isn't it cached in the downloads dir on the builder?
Dec 04 00:38:14 <ka6sox> don't think so...its been failing there.
Dec 04 00:38:29 <mwester> hmm... ok.
Dec 04 00:38:45 <ka6sox> I need the version specifically that slugOS uses.
Dec 04 00:38:50 <mwester> Which SlugOS version?
Dec 04 00:38:54 <ka6sox> (slugOS5 I believe
Dec 04 00:38:55 <ka6sox> )
Dec 04 00:39:01 <ka6sox> jas...its on the ML.
Dec 04 00:39:32 <mwester> Ok, I shall look about on my build machine when I get home (still travelling).
Dec 04 00:40:18 <mwester> Hmm.. i may have a VM image with an old downloads dir on it.  I might be able to get at that machine tonight later.
Dec 04 00:40:34 <mwester> BTW, you are sure it's ipkg-native, and not opkg?
Dec 04 00:40:44 <ka6sox> pkg-utils-native-1.6+cvs20050404-r25
Dec 04 00:40:57 <ka6sox> +i]
Dec 04 00:41:02 <ka6sox> ipkg-utils-native-1.6+cvs20050404-r25
Dec 04 00:42:27 <mwester> Ah, ok.  Not sure how we handle CVS, IIRC bb needs to check with the original repo unless we nailed down a specific version or date from the repo.
Dec 04 00:43:46 <ka6sox> mwester, sent a pastebin with output
Dec 04 00:43:55 <mwester> thx.
Dec 04 00:44:34 <ka6sox> mwester, I'm trying to bring our mirror up to date here and find lost sources.
Dec 04 00:45:22 <mwester> Got it.
Dec 04 00:46:04 <mwester> We may need special handling for that one, then.  I'll work on it... I also seem to recall that we should be switching to opkg-native instead, but not sure if/how we do that.
Dec 04 00:46:34 <ka6sox> iirc the okpg maintainer is here.
Dec 04 00:48:01 <mwester> yep
Dec 04 00:48:38 <mwester> but if possible we should find the ipkg sources, for no other reason than to make old versions of slugos buildable.
Dec 04 00:51:22 <risca> When writing a kernel recipe, which of these 2 options is preferred: do_configure() { oe_runmake ea3250_defconfig } or do_configure_prepend() { oe_runmake ea3250_defconfig } ?
Dec 04 00:51:53 <risca> I've seen both methods in different recipes
Dec 04 00:53:16 <ka6sox> mwester, thanks for your help
Dec 04 00:57:42 <ka6sox> risca, I've seen both too.
Dec 04 00:57:46 <ka6sox> dunno which is right
Dec 04 00:58:05 <risca> I think I'll go with do_configure
Dec 04 00:58:18 <risca> It is a configuration step after all
Dec 04 01:40:29 <khem> risca: do
Dec 04 01:40:56 <khem> do_configure_append will add the code to existing do_configure
Dec 04 01:41:37 <risca> khem, Thanks for the tip =)
Dec 04 01:48:47 <gregoiregentil> in a recipe, what's the variable name for "i686-linux". I tried various combinations around HOST but it's not giving what I want. Any idea?
Dec 04 01:50:20 <kergoth> gregoiregentil: HOST_SYS if HOST_VENDOR is unset, otherwise HOST_ARCH + HOST_OS
Dec 04 01:50:32 <gregoiregentil> trying...
Dec 04 01:51:26 <kergoth> khem: heh, it was only publicly announced the last day of the summit, couldn't until then :)
Dec 04 01:53:13 <gregoiregentil> kergoth: it's what I tried and HOST_ARCH _ HOST_OS gives me arm linux-gnueabi
Dec 04 01:53:27 <gregoiregentil> it's pretty far from i686-linux ;-)
Dec 04 01:53:31 <kergoth> well, you didn't tell me you were targeting arm, did you?
Dec 04 01:53:34 <kergoth> no, you didn't
Dec 04 01:53:41 <kergoth> what you want is the arch of your *build* machine
Dec 04 01:53:41 <gregoiregentil> no I didn't ;-)
Dec 04 01:53:43 <kergoth> see BUILD_SYS
Dec 04 01:53:52 <gregoiregentil> yes. trying again!
Dec 04 01:53:54 <kergoth> BUILD, HOST, and TARGET are the existing vars
Dec 04 01:54:07 <kergoth> HOST is either build or target, depending on the recipe
Dec 04 01:54:13 <kergoth> (native, cross, ..)
Dec 04 01:54:54 <gregoiregentil> I understand now!
Dec 04 01:56:01 <kergoth> :)
Dec 04 01:56:20 <kergoth> see openembedded/conf/bitbake.conf for many of the main variables with their default values
Dec 04 01:56:26 <gregoiregentil> working
Dec 04 01:56:32 <gregoiregentil> Thanks
Dec 04 01:56:56 <gregoiregentil> On another note, I'm trying to cross compile scrollkeeper. I get the following error:
Dec 04 01:56:56 <gregoiregentil> /work/OE/build/tmp/work/armv7a-angstrom-linux-gnueabi/scrollkeeper-0.3.14-r0/scrollkeeper-0.3.14/buildtools/.libs/lt-scrollkeeper-tree-separate: error while loading shared libraries: libdl.so.2: cannot open shared object file: No such file or directory
Dec 04 01:57:10 <gregoiregentil> lt-scrollkeeper-tree-separate is arm according to readelf
Dec 04 01:57:14 <gregoiregentil> and I have
Dec 04 01:57:23 <gregoiregentil> /OE/build/tmp/sysroots/armv7a-angstrom-linux-gnueabi/lib/libdl.so.2
Dec 04 01:57:41 <gregoiregentil> so what do I need to do so that the shared libraries is found?
Dec 04 01:59:47 <gregoiregentil> or should I compile native first and modify the recipe so that it's using some binaries from the native package?
Dec 04 02:59:17 <zenlinux> I see the release was tagged. Nice work everyone!
Dec 04 02:59:35 * zenlinux throws some virtual confetti
**** ENDING LOGGING AT Sat Dec 04 02:59:59 2010